Gas Turbine Combustion Network Analysis for Instability Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Gas turbines experience intermittent malfunctions due to thermoacoustic, aeroacoustic, and aeroelastic instabilities, which can lead to damage and reduced performance, and existing methods often require trial-and-error approaches and design changes, making them inefficient and risky.
Innovation Solution
A computer-implemented method that generates a digital map of the combustion volume, analyzes state variables using sensors or simulations, and forms a network to identify instability patterns, allowing for controlled disruption to minimize these instabilities through active or passive control.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If trial-and-error procedures are used to control thermoacoustic instabilities, then design changes can be applied to resolve instabilities, but the process requires a large number of attempts and is time-consuming
Solution Approach 1:
The patent applies preliminary action by using digital simulations and network analysis to identify critical regions and instability patterns before actual gas turbine operation. The method pre-determines optimal control element positions and disruption strategies through virtual modeling, allowing engineers to predict and prevent thermoacoustic instabilities without requiring extensive trial-and-error testing on physical hardware.
Solution Approach 2:
The patent replaces traditional mechanical trial-and-error testing with a computational approach. Instead of physically testing design changes through repeated attempts, the system uses digital simulations, sensor data analysis, and network theory to model combustion processes and identify instability mechanisms, substituting physical experimentation with virtual analysis and prediction.
2Reliability
If controlled disruption is applied to minimize instabilities, then thermoacoustic instabilities can be reduced, but the combustion process must be deliberately perturbed
Solution Approach 1:
The patent applies local quality by targeting controlled disruption to specific critical regions within the combustion volume rather than applying uniform disruption throughout. The network analysis identifies particular zones where instability originates or propagates most strongly, and control elements are positioned and activated only in those localized areas, minimizing overall combustion disruption while effectively suppressing instabilities.
Solution Approach 2:
The patent introduces control elements as intermediaries between the combustion process and the instability phenomenon. These control elements (such as acoustic actuators or flow control devices) serve as mediators that apply targeted disruptions to interrupt instability feedback loops without directly interfering with the overall combustion process, thereby reducing instabilities while maintaining combustion efficiency.
3Reliability
If design changes are applied retrospectively to gas turbines, then instability problems can be addressed, but the turbine must be taken offline and modified
Solution Approach 1:
The patent implements feedback by using sensors to continuously monitor combustion parameters and instability indicators during gas turbine operation. This real-time data feeds into the network analysis system, which identifies emerging instability patterns and triggers controlled disruption only when needed. This feedback-based approach allows the turbine to operate normally without modifications until instabilities occur, at which point automated control actions suppress them without requiring shutdown or physical redesign.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a computer-implemented method for minimizing thermoacoustic, aeroacoustic and/or aeroelastic instability states of a combustion process in a gas turbine, wherein the gas turbine has a combustion volume comprising a fluid, wherein the steps of the method involve the following: generating a digital model of the combustion volume in the form of a mesh by spatially dividing the combustion volume into volume regions and digitally modelling these volume regions using mesh cells; generating a temporal sequence of at least one state variable of the fluid contained in each volume region during a combustion process, by digital simulation of the combustion process and/or measurement of the state variable during the combustion process using at least one sensor; applying the temporal sequence to each mesh cell; building a spatial and/or temporal network in which each mesh cell is represented by a node and node pairs correlated with respect to the temporal sequences are connected by a link; carrying out a network analysis to determine at least one statistical network measure of the network, carrying out an analysis of the network measure, wherein the analysis of the network measure is configured for minimizing thermoacoustic instability states of a combustion process by controlled disruption of the combustion process using at least one control element. The invention further relates to a device for carrying out the method.